The problem seems to be caused by this fix:
2006-08-20 Richard Stallman <[EMAIL PROTECTED]>
* emacs-lisp/timer.el (run-with-idle-timer): Pass t to
timer-activate-when-idle, so timer can run before Emacs becomes
non-idle again.
After reverting that fix, I see Emacs CPU usage go from 0 to 2% for a
brief instant when the idle timers kick in. With that fix, the CPU usage
goes to 50% and stays there until I perform an input event to break the
idle loop.
It is not surprising that that change would have this effect
if a program uses the function this way.
I made the change because it seems to be a bug that the timer would
not run until after Emacs was non-idle once again. If this change
causes problems, I can revert it. However, the current (changed)
behavior is more useful in general, so I would want to introduce a new
way to invoke the current behavior.
I wonder why semantic is doing this. Could it do the job
by having a timer that repeats automatically?
_______________________________________________
emacs-pretest-bug mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/emacs-pretest-bug
